Osteogenesis Imperfecta (OI) is a complex group of disorders characterized by excessive fragility of bone and pathologic fractures. Substantial evidence has now accumulated that most cases of OI are due to defects in the expression or structure of Type I collagen, the principal collagen of bone. Of the approximately 25 cases associated with structurally abnormal collagen chains, the majority are single amino acid substitutions for glycine residues. Most of the defined defects have occurred in lethal OI; non-lethal OI has received little study although biochemically such cases appear to be similar to lethal OI. Despite this progress, the pathogenesis of OI is essentially unknown and it is entirely unclear why some mutations produce lethal OI whereas similar mutations elsewhere produce mild disease. Current notions of pathogenesis focus on the collagen molecule (rather than the fibril), and none provides adequate explanation for all of the abnormalities found in OI. In this proposal, we present a coherent hypothesis for the pathogenesis of OI-the bone collagen fibril structure-function hypothesis-which focuses on the supramolecular structure of bone collagen, the fibril. This hypothesis suggests that there exist specific functional domains on the surface of bone Type I collagen fibrils, that such domains mediate interactions with other bone matrix components critical for osteogenesis, and that structural defects of constituent collagen molecules impairs this function, producing defective bone. This hypothesis predicts the existence of sets of homologous mutations, related by a simple D-stagger periodicity, which produce similar clinical phenotypic effects, ie. mild, moderate, severe or lethal OI. The hypothesis also provides explanation for the observed deficient content of specific bone phosphoproteins found in OI bone. We propose to test this hypothesis by defining the molecular defects in multiple cases of OI, mapping such mutations onto the collagen fibril, and correlating map position with clinical phenotype. The resultant phenotype- fibril domain map is anticipated to support the hypothesis and provide critical justification for interventional studies designed to prove the hypothesis. These data would provide substantial insight into the pathogenesis of OI, define the structure-function relationships of bone collagen fibrils, and contribute to understanding of normal bone formation.